Text vs. Graphical Links in an E-commerce Website

Discussion

It was expected that the treatment of the instructions in both graphics and text would be the quickest and the most error free. However after running through the trials, the group found that the textual links were less likely to cause an error. This could mean one of two things the pictures were misleading, or people did not take their time to read the text captions. In our usability survey however, people tended to say that the textual and graphic links were much easier to understand.

When subjects used the treatment with both graphics and text links, the users tended to click on the underlined text rather than the graphic to move to the next page. This could be due to the fact that many subjects, used to using the web, reconized the underlined text as a hyperlink. Where as, in a lot of cases on the web, pictures are not hyperlinks, but just graphics. The fact that the subjects are not used to pictures being hyperlinks could have caused users to take so much time on the graphical treatment.

Another interesting occurrence was on the graphical treatment, 14 out of 24 subjects visited at least one extra page when finding the desired product. This could be due to the fact that the labels on the pages were misleading or the subjects thought they clicked on the wrong image and did not take their time to make sure.

Another aspect dealing with errors made by subjects was the fact that once a mistake was made, it took users a while to realize that they had made a mistake, a good 10-15 seconds. This could be due to users entering in the incorrect price and getting an error message, thus taking time. Another possible reason for the time to pick up the error could be due to users taking a while to understand that they needed to go back and make another selection.

Along with the time differences, subjects tended to make mistakes with the graphical links of washers and dryers. This could be due to the fact that the images looked almost identical. Also users tended to do one of two things when stumped with a graphical link, guess right a way, or take a long time before making a selection, there really was no middle ground in selecting a link.

An interesting side note is that when presented with a composite image of a group of merchandise, subjects usually clicked on the part of the image with the merchandise they were looking for even though the link was connected to the whole picture. Though this aspect was not measured, it was noted through members of the group administering the study.

In addition to the position of the user clicking on a link, users also tended to click on the top left link when they where confunsed, and then circle around until they selected the correct product. The group had expected subjects to try and pick up visual cues from the products, company names or some other distingising feature. However,users tended to rather make a mistake, then study every link until they found a identifing feature.

Department of Computer Science: Direct questions and comments to the student editorial team